Diagnosing a profit decline: the reference case
Two full worked profitability cases from prompt to recommendation, one interviewer-led and one candidate-led, the standard the other case types follow.
Key takeaways
- In many offices, interviewer-led versions (common at McKinsey) hand you the prompt and the exhibit and ask pointed questions in order.
- The strong candidate read the problem (flat revenue means a cost problem), narrowed fast, and asked for the facts it needed.
- Clarify the goal and the period, then state a hypothesis early.
- Split profit into revenue and cost first, always.
- Use flat or moving revenue to decide whether to start with revenue or cost, but remember that flat revenue can hide price and volume moving in opposite directions.
What this case type is and when it shows up
Profitability is the most common case type, and being good at it carries you a long way. The setup is almost always the same: a company's profit has fallen, or is lower than the client wants, and you need to find out why and what to do. Because it is so common, interviewers have seen every memorized framework applied to it, so this is where tailoring beats reciting.
Recall the core idea from the structuring chapter. Profit is revenue minus cost. So if profit fell, either revenue dropped or cost rose (or both), and that split is your first move every time. From there you narrow: revenue splits into price, volume, and mix (the balance between cheaper and dearer products), and cost splits into fixed and variable. You keep narrowing until you reach the driver that moved.
Two tools take you deeper. Contribution per unit is price minus variable cost per unit: what each sale adds toward fixed costs and then profit. Break-even volume is fixed costs divided by contribution per unit. And when revenue changes, split it: the volume effect is the change in units times the old price, and the price effect is the change in price times the new units. The second worked case below uses all three.
- Profit has fallen
- Revenue fell
- Price per cup fell
- Cups sold fell (fewer customers, or fewer cups each)
- Mix shifted toward cheaper items
- Other revenue fell (food, packaged beans)
- Key: Cost rose
- Fixed costs rose (rent, salaries)
- Key: Variable costs rose (beans, milk, cups)
A profit tree is a tool. Tailor the branches to the business in front of you, and state which branch you expect to matter.
The reference case: the Harqeen Coffee chain
Worked case
Harqeen Coffee: profit down 20 percent
The prompt
Harqeen Coffee is a coffee chain with 150 stores in the UAE and Saudi Arabia. Last year it made AED 10 million of profit on revenue of AED 100 million. This year revenue was again AED 100 million, but profit fell to AED 8 million. The chief executive wants to know why and what to do.
An interviewer-led profitability case with an exhibit. Watch how the revenue-versus-cost split, one exhibit, and two well-placed questions narrow the problem to a single driver.
Clarifying questions, with the interviewer's answers
- What is the client's goal?Answer: Understand the fall and get profit back to AED 10 million next year.
- Over what period, and did anything unusual happen, such as store openings or closures?Answer: This year versus last year, same 150 stores.
- Are competitors seeing the same fall?Answer: Their profits are also under pressure, but we have no detail yet.
A hypothesis to say out loud: Revenue is flat, so the fall must come from cost. Coffee is a traded commodity whose price moves a lot, so my hypothesis is that an input cost, most likely coffee beans, has risen. I will split cost to test it.
The structure
- Profit fell on flat revenue, so cost rose. Which cost, and why?
- Revenue is flat at AED 100 million: a cost problem
- Split cost into buckets and find the one that moved
- Key: Split that bucket into volume, recipe, and input price
The exhibit
Grouped bar chart: Harqeen Coffee cost breakdown, last year versus this year. Values in AED millions. Series: Last year; This year. Rent: Last year 14, This year 14; Wages: Last year 30, This year 30; Coffee beans: Last year 8, This year 10; Milk and other ingredients: Last year 12, This year 12; Other (utilities, marketing, admin): Last year 26, This year 26.
Working it through
1. Size the problem
Profit fell from AED 10 million to AED 8 million, a drop of AED 2 million, which is a 20 percent fall. Profit margin went from 10 percent to 8 percent.
Profit fall (AED m):10 - 8 = 22. Confirm the cost rise
Cost is revenue minus profit: AED 90 million last year and AED 92 million this year. Cost rose by the same AED 2 million that profit fell.
Cost rise (AED m):(100 - 8) - (100 - 10) = 23. Read the exhibit
The cost breakdown shows rent, wages, milk and other ingredients, and other costs flat. Only coffee beans rose, from AED 8 million to AED 10 million. That rise is the entire profit drop.
Bean cost rise (AED m):10 - 8 = 24. Ask, do not assume
Bean cost is cups sold times grams of coffee per cup times price per kilogram. Flat revenue does not prove that cups sold are flat: a price rise could hide fewer cups. So the candidate asks. Ask: Did the average price per cup or the number of cups change? Answer: Both are flat. Ask: Did the recipe or cup size change? Answer: No.
5. Find the input-price rise
Cups and grams per cup are unchanged, so the whole rise in bean cost is the price per kilogram, up about 25 percent. This is an input-price problem, not a demand or pricing problem.
Bean price rise (%):(10 - 8) ÷ 8 × 100 = 256. Check the market
Ask: Are competitors facing the same bean prices? Answer: Yes, world coffee prices rose and rivals report the same pressure. That matters: if rivals face the same cost, a small price rise is less likely to send customers to them.
7. Size a price response
To recover AED 2 million on AED 100 million of revenue with the same number of cups, the average price must rise about 2 percent.
Price rise needed (%):2 ÷ 100 × 100 = 28. How many cups can the price rise lose?
Variable costs are beans (AED 10 million) plus milk and other ingredients (AED 12 million), so contribution is AED 78 for every AED 100 of revenue. After a 2 percent price rise it is AED 80 for the same cups. The rise still adds profit unless cups fall by more than about 2.5 percent.
Break-even fall in cups (%):(1 - 78 ÷ 80) × 100 = 2.5
What the exhibit shows
Every cost bucket is flat except coffee beans, which rose AED 2 million. Since profit fell by AED 2 million, rising bean cost explains the whole fall.
The recommendation
The whole AED 2 million profit fall comes from coffee-bean prices, so the fix is to recover that cost, not to chase demand. First, every cost bucket except beans is flat, and bean cost rose AED 2 million, about 25 percent, with cups, prices, and the recipe unchanged. Second, competitors face the same bean prices, so a small price rise is realistic: about 2 percent on average recovers the full AED 2 million if cups hold. Third, that rise still adds profit unless cups fall by more than about 2.5 percent. I recommend three actions: fix part of next year's bean cost through longer supplier contracts or hedging (agreeing a price now for future purchases); pilot a 2 percent price rise in a few stores, focused on drinks where customers are least price-sensitive, and watch cup volume; and test blend or sourcing changes that protect taste at a lower cost.
Risks: A price rise could cut cups sold by more than 2.5 percent if customers are more price-sensitive than expected; Locking in bean prices could cost more than the market if world prices fall; A new blend or supplier could change the taste customers expect.
Next steps: Open supplier and hedging discussions this month; Pilot a 2 percent price rise in 10 stores for six weeks and track cups sold.
A strong candidate
Stated a hypothesis early, used flat revenue to go straight to cost, used the exhibit to isolate beans, asked instead of assuming that cups and recipe were flat, checked competitors, and closed with a quantified recommendation, its risk, and a pilot.
A weak candidate
Opened by reciting the three Cs and the four Ps, spent time on marketing ideas, assumed cups were flat without asking, and recommended "raising prices" with no size and no risk.
Prompt: "Harqeen Coffee's profit fell 20 percent on flat revenue."
Weaker answer
Says: "I will look at the customer, the company, the competitor, then the four Ps." Spends ten minutes on demand and marketing while the answer, a flat revenue line pointing at costs, is in plain sight.
Stronger answer
Says: "Revenue is flat, so this is a cost problem. My hypothesis is an input cost such as beans. Let me split cost and find which bucket moved." Uses the exhibit to land on beans, asks whether cups and the recipe changed, finds a 25 percent bean-price rise, and recommends hedging plus a piloted 2 percent price rise.
Why the stronger answer wins: The strong candidate read the problem (flat revenue means a cost problem), narrowed fast, and asked for the facts it needed. The weak candidate forced a generic framework onto it and never engaged the facts.
Methods for working any profitability case
- Clarify the goal and the period, then state a hypothesis early.
- Split profit into revenue and cost first, always.
- Use flat or moving revenue to decide whether to start with revenue or cost, but remember that flat revenue can hide price and volume moving in opposite directions.
- Narrow the side that moved: revenue into price, volume, and mix; cost into fixed and variable, then into specific buckets.
- When a step needs a fact, ask the interviewer rather than assume it.
- Check competitors: a cost rise that hits the whole industry allows different fixes than one that hits only your client.
- Quantify the driver and the fix, so your recommendation rests on numbers.
Common mistakes, traps, and curveballs
- Reaching for a generic framework instead of the revenue-versus-cost split.
- Digging into revenue when revenue is flat and the answer is in cost.
- Assuming price and volume are flat because revenue is flat.
- Reading every exhibit number aloud without stating the so-what.
- Stopping at "costs rose" without finding which cost and why.
- A curveball where both revenue and cost moved, so you must size each effect (see the next worked case).
Curveball: revenue and cost both moved
Worked case
Brenholt: price up, volume down, fixed costs up
The prompt
Brenholt, a bakery chain in Germany, saw profit fall from EUR 4 million to EUR 2.85 million. Last year it sold 10 million loaves at EUR 3.00; this year it sold 9.5 million at EUR 3.10. Variable cost is EUR 1.80 per loaf in both years. Fixed costs rose from EUR 8 million to EUR 9.5 million. What drove the fall?
Candidate-led: you drive the analysis and ask for what you need; the interviewer answers only what you ask.
Clarifying questions, with the interviewer's answers
- Is variable cost per loaf the same in both years?Answer: Yes, EUR 1.80.
- What caused the rise in fixed costs?Answer: Three new stores opened late in the year.
A hypothesis to say out loud: Both revenue and cost moved, so I will size each effect. My hypothesis is that the higher fixed costs, not the lower volume, drive most of the fall, because the price rise offsets much of the lost volume.
The structure
- Split the profit change into revenue effects and cost effects
- Revenue change: volume effect and price effect
- Contribution change (price minus variable cost, times loaves)
- Key: Fixed-cost change
Working it through
1. Revenue change
Candidate: "I will start with revenue: 9.5 million loaves at EUR 3.10 this year against 10 million at EUR 3.00 last year." Revenue fell only EUR 0.55 million, but profit fell EUR 1.15 million, so something else moved as well.
Revenue change (EUR m):9.5 × 3.1 - 10 × 3 = -0.552. Volume effect
The change in loaves times last year's price.
Volume effect (EUR m):(9.5 - 10) × 3 = -1.53. Price effect
The change in price times this year's loaves. Volume and price effects add up to the revenue change.
Price effect (EUR m):(3.1 - 3) × 9.5 = 0.954. Contribution, last year
Candidate: "The price rise carries no extra variable cost, so contribution may tell a different story from revenue. Is variable cost per loaf really unchanged?" Interviewer: "Yes, EUR 1.80 in both years." Contribution is loaves times (price minus variable cost).
Contribution last year (EUR m):10 × (3 - 1.8) = 125. Contribution, this year
Contribution rose even though revenue fell, because the price rise goes straight to contribution on every loaf.
Contribution this year (EUR m):9.5 × (3.1 - 1.8) = 12.356. Profit bridge
Candidate: "Contribution rose, so the fall must come from fixed costs." Interviewer: "Why does that matter for the client?" Candidate: "Because the fix is different: this is about the new stores, not about demand or price." Change in contribution minus change in fixed costs equals the change in profit.
Profit change (EUR m):(12.35 - 12) - (9.5 - 8) = -1.15
The recommendation
The EUR 1.15 million fall in profit comes from the three new stores' fixed costs, not from weak demand. First, revenue fell EUR 0.55 million, but the price rise more than made up for the lost loaves at the contribution level. Second, contribution actually rose, by EUR 0.35 million, because the price rise carries no extra variable cost. Third, fixed costs rose EUR 1.5 million with the three new stores, which explains the whole fall. Next, check whether the new stores are on track to cover their fixed costs, and find out why total loaves fell 5 percent even with more stores.
Risks: The new stores may need longer than a year to reach normal sales; Further price rises could cost more volume than this year's did.
Next steps: Track weekly sales and contribution at each new store; Compare loaves per store in old stores this year and last year.
A strong candidate
Split revenue into price and volume, used contribution rather than revenue, and found that fixed costs drove the fall.
A weak candidate
Saw revenue fall and recommended a marketing campaign to win back volume, missing that contribution had risen.
In many offices, interviewer-led versions (common at McKinsey) hand you the prompt and the exhibit and ask pointed questions in order. Candidate-led versions (common at Bain and BCG) often expect you to drive: lay out the tree, ask for the cost breakdown yourself, and steer to the driver. The diagnosis is the same; who leads differs. Formats differ by office and change over time, so check the current process for your target office.
Practice
A company has revenue of 100 and profit of 8. Last year profit was 10 on the same revenue. By what percent did profit fall?
A cafe in Singapore sells coffee at SGD 5 a cup with a variable cost of SGD 2 a cup. Fixed costs are SGD 30,000 a month. How many cups a month does it need to break even?
Units fell from 200,000 to 180,000 while price rose from GBP 10 to GBP 11. What is the price effect on revenue, in GBP (change in price times this year's units)?
Harqeen Coffee's revenue is flat but profit fell. What is the strongest first move?
Revenue fell 5 percent, but contribution rose. What most likely happened?
A retailer's profit fell. Revenue is up 8 percent, but it opened 15 percent more stores. What should you check before calling revenue healthy?
Split profit into revenue and cost, narrow to the one driver that moved, ask for any fact you would otherwise assume, and end with a recommendation that has numbers, a risk, and a next step.
Sources for this lesson (1)
- Recognized public explanations of case-interview concepts and frameworks
My notes on this lesson
0 of 5,000 characters. Saves automatically.
Try the 6 remaining checks and drills above to complete this lesson (0 of 6 done).
Spotted something wrong or out of date? Report a mistake. We check every report and correct the page.